Biochemistry 1992, 31, 3935-3940 Binder, A., Jagendorf, A., & Ngo, E. (1978) J. Biol. Chem. 253, 3094-3 100. Boyer, P. D. (1987) Biochemistry 26, 8503-8507. Boyer, P. D., & Kohlbrenner, W. E. (1981) in Energy Coupling in Photosynthesis (Selman, B. R., & Selman-Reiman, S., Eds.) pp 23 1-240, Elsevier/North-Holland, New York. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254. Bruist, M. F., & Hammes, G. G. (1981) Biochemistry 20, 6298-6305. Cerione, R. A., & Hammes, G. G. (1982) Biochemistry 21, 745-752. Duncan, T. M., Parsonage, D., & Senior, A. E. (1986) FEBS Lett. 208, 1-6. Forster, T. (1959) Discuss. Faraday SOC.27, 7-17. Houmard, J., & Drapeau, G. R. (1972) Proc. Natl. Acad. Sci. U.S.A. 69, 3506-3509. Leckband, D., & Hammes, G. G. (1987) Biochemistry 26, 2306-2312. Lien, S., & Racker, E. (1971) Methods Enzymol. 23a, 547-555. Maggio, M. B., Pagan, J., Parsonage, D., Hatch, L., & Senior, A. E. (1987) J. Biol. Chem. 262, 8981-8984. McCarty, R. E., & Moroney, J. V. (1985) in The Enzymes of Biological Membranes (Martonosi, A., Ed.) 2nd ed., pp 383-413, Plenum Press, New York. McCarty, R. E., & Hammes, G. G. (1987) Trends Biochem. Sci. 12, 234-237. Mills, D. A., & Richter, M. L. (1991) J. Biol. Chem. 266, 7440-7444. Nalin, C. M., & Nelson, N. (1987) Curr. Top. Bioenerg. 15, 273-294.
3935
Parker, C. A., & Reese, W. T. (1966) Analyst (London) 85, 5 87-600. Parvan, R., Pecht, I., & Soreq, H. (1983) Anal. Biochem. 133, 450-455. Penefsky, H. S . (1977) J. Biol. Chem. 252, 2891-2899. Penefsky, H. S., & Cross, R. L., (1990) Adv. Enzymol. Relat. Areas Mol. Biol. 64, 173-214. Richter, M. L., & McCarty, R. E. (1987) J. Biol. Chem. 262, 15037-1 5040. Richter, M. L., Patrie, W. J., & McCarty, R. E. (1984) J . Biol. Chem. 259, 7371-7373. Richter, M. L., Snyder, B., McCarty, R. E., & Hammes, G. G. (1985) Biochemistry 24, 5755-5163. Richter, M. L., Gromet-Elhanan, Z., & McCarty, R. E. (1986) J. Biol. Chem. 261, 12109-12113. Scott, T. G., Spencer, R. D., Leonard, M. J., & Weber, G. (1970) J . Am. Chem. SOC.92,687-695. Shapiro, A. B., & McCarty, R. E. (1988) J. Biol. Chem. 263, 14 160-1 4 165. Sippel, T. 0. (1981) J. Histochem. Cytochem. 29,1377-1381. Snyder, B., & Hammes, G. G. (1984) Biochemistry 23, 5787-5795. Snyder, B., & Hammes, G. G. (1985) Biochemistry 24, 2324-2331. Walker, J. E., Saraste, M., & Gay, N. J. (1984) Biochim. Biophys. Acta 768, 164-200. Walker, J. E., Fearnley, I. M., Gay, N. J., Gibson, B. W., Northrop, F. D., Powell, S . J., Runswick, M., Saraste, M., & Tybulewicz, V. L. T. (1985) J. Mol. Biol. 184,677-701. Xue, Z., Zhou, J.-M., Melese, T., Cross, R. L., & Boyer, P. D. (1987) Biochemistry 26, 3749-3753.
Colchicine Photosensitizes Covalent Tubulin Dimerization J. Wolff,* Jean Hwang, Dan L. Sackett, and Leslie Knipling Laboratory of Biochemical Pharmacology, National Institute of Diabetes and Digestive and Kidney Diseases. National Institutes of Health, Bethesda, Maryland 20892 Received November 13, 1991 ABSTRACT: Pure rat brain tubulin can be cross-linked by ultraviolet irradiation of tubulin-colchicine complexes at the high-wavelength maximum of colchicine to form covalent dimers > trimers > tetramers. With colchicine concentrations -3 X lo4 M (mole ratio to tubulin 3-12) and irradiation for 5-10 min at 95-109 mW/cm2, the yield of dimers is 11-17% and of trimers is 4-6% of the total tubulin. The oligomers show polydispersity and anomalously high apparent molecular masses that converge toward expected values in low-density gels. Maximal dimer yields are obtained with MTC and the decreasing photosensitizing potency is MTC > colchicine > colchicide > isocolchicine > thiocolchicine. Single-ring troponoids also promote dimerization. Evidence is presented suggesting that the initial, low-affinity, binding step of colchicine and its analogues is sufficient to photosensitize tubulin dimerization.
C o v a l e n t labeling of tubulin with [3H]colchicine can be accomplished by ultraviolet irradiation of the tropolone moiety of the drug at -353 nm (Wolff et al., 1991). The label is found primarily in the 8-monomer, but some labeling of the a-subunit can be induced under mild denaturing conditions suggesting that the drug might span the a l p subunit interface. This suggested that irradiation of bound colchicine might covalently cross-link the two subunits, and we have investigated this possibility. Earlier studies, employing various chemical cross-linking agents, had shown that covalent tubulin oligomers could be formed in good yield. Thus, Luduena et al. (1977),
using chick brain tubulin, showed that a@, as well as homologous dimers, could be identified by mobility and double-isotope methods; Galella and Smith (1982) found that oligomers up to hexamers were formed from tubulin treated with various bifunctional imido esters having spacers of 5-10 A. Subsequently, it was found that hexanedione cross-linked tubulin to dimers and larger forms (Boekelheide, 1987; Sioussat & Boekelheide, 1989). Ultraviolet irradiation of tubulin at low wavelengths has long been known to lead to destruction of mitotic spindles, etc., which is thought to result from conformational changes in
This article not subject to US.Copyright. Published 1992 by the American Chemical Society
3936 Biochemistry, Vol. 31, No. 16, 1992 tubulin as shown by circular dichroic and ultraviolet spectra, loss of SH groups, and cross-linking (Zaremba et al., 1984). However, the covalent dimerization we report here results from irradiation of the tropolone moiety of colchicine at wavelengths that are not directly absorbed by proteins without colchicine. Irradiation without colchicine at these wavelengths does not cause dimerization. Such effects can, therefore, be considered to be colchicine-sensitized photodimerization of tubulin. EXPERIMENTAL PROCEDURES Methods. Rat brain tubulin was prepared by temperature-dependent polymerization-depolymerization cycles followed by treatment with 1.0 M sodium glutamate (Hammel & Lin, 1984). It was >98% pure in overloaded SDS’ gels and was kept in liquid N2until just before use. Residual guanidine nucleotides were not removed, but all subsequent reactions were carried out in the absence of GTP to prevent polymerization. Except where noted, mixtures were incubated in assembly buffer (100 mM MES, pH 6.85, 1 mM MgC12, 1 mM EGTA) at 37 “Cin the dark for 30-60 min and were then irradiated at 4 OC by a water-cooled @ram high-pressure mercury lamp at 90-100 W of lamp power. Delivered power was measured with a Spectroline DM365N radiometer and a neutral filter. To protect the protein and to filter out infrared components, all irradiations were carried out under 2 cm of a 20% CuS04-5H20solution (Wolff et al., 1991; Kasha, 1949) which permitted < I % transmission at 305 nm, 50% transmission at 322 nm, and 97% transmission at 353 nm, the absorption maximum of colchicine. The samples were irradiated in cooled polyethylene Eppendorf tubes, generally in a volume of 20 pL. Irradiation times were varied as described in the text. Samples were then boiled for 2-3 min in an SDS/mercaptoethanol loading solution, cooled, and electrophoresed on 8% SDS/polyacrylamide gels (bonded on GelBond sheets unless otherwise noted) at 20 mA until 15-20 min beyond the dye-expulsion time, unless otherwise noted. In some experiments the mercaptoethanol was deleted. After being stained and destained, gels were soaked in 5% glycerol in water for 6-15 h, dried, and covered with a plastic sheet. They were then scanned in a Beckman Model DU-8 gel scanner, and the machine-computed areas were used to calculate the percent of distribution on each lane. The linearity of the scanner response was checked for the range of the rather high protein concentrations used in this study. The response was linear &lo% (expressed as the fraction present as dimers) over a range of 10-50 pg of total protein per lane. Moreover, several gels were rescanned after one year and yielded identical resu 1t s. Transverse gradient gels were prepared by forming linear 4-996 polyacrylamide gradients at right angles to the running direction. Stacking and loading gels were then applied as usual, and alternate sample and standard lanes were electrophoresed across the whole gradient to permit analysis according to Ferguson (Andrews, 1986). Materials. Commercial chemicals were of the highest purity available. MTC [2-methoxy-5-(2’,3’,4’-trimethoxyphenyl)tropone] was synthesized by Dr. T. Fitzgerald, Flordia A&M University. Colchicoside was a gift of Dr. C. Chignell; colchicide, thiocolchicine, and isocolchicine were generously provided by Dr. Arnold Brossi of our institute. Tropolone was from Aldrich, and 2-methoxy-5-nitrotroponewas from American Tokyo Kasai. Abbreviations: MTC, 2-methoxy-5-(2’,3’,4’-trimethoxyphenyl)tropone; SDS, sodium dodecyl sulfate; EGTA, ethylene glycol bis(8aminoethyl ether)-N,N,N’,N’-tetraacetic acid.
Wolff et al.
T
200 116 97 66
45
12
-
-
D
-
r l J l l l l I
I
5
4
-
3
2
COLCHICINE [-log M] FIGURE 1: The effect of colchicine concentration on dimer formation by irradiation of rat brain tubulin. (A) A 50 p M concentration of rat brain tubulin was incubated with colchicine for 30 min at 37 OC, irradiated for 5 min at 109 mW/cm2, and electrophoresed on an 8% gel. Colchicine concentrations (millimolar) are listed above each lane. a and /3 stand for the tubulin monomers; D and T stand for the presumptive dimers and trimers. Numbers refer to marker proteins in kilodaltons. (B) Densitometric scans of above Coomassie-stained gel.
RESULTS Under the present incubation conditions, reversibly bound colchicine amounts to 0.7-0.8 mol/mol of dimer (Bhattacharyya & Wolff, 1974). However, irradiation for covalent binding conditions is competed for by photoisomerization of the colchicine to lumicolchicine and, hence, yields are expected to be much lower (Wolff et al., 1991). Moreover, the nature of the binding reaction for dimerization may not be entirely the same as for formation of the covalent colchicinetubulin adduct. Thus, the correlation between reversible or covalent binding and dimerization is difficult to quantify. Nevertheless, many other aspects of the photosensitized dimerization of tubulin can be studied. The yield of dimers (and higher oligomers) depends on the concentration of colchicine present during irradiation, attaining a maximum at (1-3) X IO4 M colchicine in various experiments when the tubulin concentration was 50 pM (that is, at a mole ratio of 4-12 for colchicine to tubulin). A typical example is depicted in Figure 1A with the corresponding scanning results for the total dimer and trimer peaks in Figure 1B. Similar maxima were observed for presumptive trimer formation; the presumptive tetramer bands were often too weak for quantitation at these protein concentrations. It should be noted that the concentrations of colchicine yielding maximal cross-linking are higher than required for saturation of the binding site. This is due to the rapid conversion of colchicine to lumicolchicine under ultraviolet light (Wilson & Friedkin, 1966). Thus, the concentration of colchicine decreases rapidly
Biochemistry, Vol. 31, No. 16, 1992 3937
Tubulin Dimerization 2
5
L
8 I-
zun
4
6
8
10 12 14 16 18 20
e Dimers
12}
8li/ 6
1
Trimers
200
- 200 - 116 - 97 - 66
-
116 -
97
- 45
-
a W ~~
2
4
6
8
10
12
14
16
IRRADIATION TIME (Mid
FIGURE2: The effect of irradiation time on photosensitized oligomer formation in tubulin by colchicine. A 30 p M conentration of tubulin and 3 X lo4 M colchicine were incubated at 37 "C for 50 min in assembly buffer and irradiated at 98 mW/cm2 for the times indicated, electrophoresed on an 8% SDS gel, and scanned.
during the period of irradiation. However, as will be shown below, these are not nonspecific effects, since a nonbinding colchicine analogue does not promote dimerization when irradiated together with tubulin. The decrease in dimer formation occurring at high colchicine concentrations is not understood; it may be partly due to the inner filter effect and also to colchicine dimer formation at the high concentrations (Engelborghs, 1981). When the colchicine/tubulin ratio was held constant at 2.4 or 4.8 with 3 X lo4 M colchicine, there was little or no effect of varying the tubulin concentration on the yield of dimer (data not shown). Three different rat brain tubulin preparations and one beef brain tubulin preparation, when irradiated with 3 X lo4 M colchicine for 10 min at 108 mW/cm2, yielded approximately 1 1-1796 of total protein as dimers and 4-6% as trimers. It is apparent from Figure 1A that the dimer (D) area is complex. With longer electrophoresis, we could invariably see five dimer bands. Gels in which disulfide bonds had not been reduced before electrophoresis contained higher percentages of protein as dimer, although they appeared as a smear (e.g., for colchicine, 10.6% dimer when reduced and 13.8% dimer when not reduced). However, the bulk of the dimer bonds was of another type. We assume that the two extra bands in addition to the expected CY& CYCY, and pp dimers arise from cross-linking of isoforms of tubulin, but this has not been established. This degree of polydispersity of dimers has also been observed in tubulin cross-linked with hexanedione (Sioussat & Boekelheide, 1989). The effect of time of irradiation at a constant power of 98 mW/cm2 is depicted in Figure 2. Maximum yields of oligomers were attained at -10 min of irradiation and then plateaued or declined slightly at longer irradiation times when the colchicine/tubulin mole ratio was 10 (Figure 2). The maximum yields of oligomers (presumptive dimer, trimer, and tetramer) amounted to -20% of the protein present on the gel. As judged by the ratios of dimer/trimer, etc., dimer formation precedes trimer formation and trimer formation precedes tetramer formation (Figure 2). Oligomer Size. Up to now, we have assumed that the next larger species present above monomer was the covalent dimer. However, from protein standards, the apparent molecular mass of this presumptive dimer was 150 kDa on 8% gels. To distinguish anomalous migration on gels from trimers, it was necessary to eliminate sieving effects (Andrews, 1986) by extrapolation of gel mobilities to zero gel densities. It is apparent from Figure 3, which depicts a 4-9% SDS transverse gradient gel (constructed at right angles to the running di-
-
FIGURE3: Transverse 4-9% SDS gradient acrylamide gel with alternating overloaded sample lanes (even numbers) and protein standard lanes at 200, 116,97,66, and 45 kDa (odd numbers). Note that the slope of the dimer "band" approaches the 116-kDa marker at 4%gel density and that the trimer peak crosses the 200-kDa marker. The a and fi monomer peaks do not separate because of overloading.
Table I: Potency of Colchicinoids in Promoting Tubulin Dimerization comwund" concn for 7% dimer vield ( M ) MTC 6 X 10" colchicine (2-4) x 10-5 8X colchicide isocolchicine 3 x IO+ 4 x 10-3 thiocolchicine 1 x 10-3 2-methoxy-5-ni trotropone 2.5 x 10-3 t romlone a Refer to Figure 4 for structures.
rection), that the slope of the line connecting the midpoints of the presumptive dimer peaks is steeper than that of the nearest marker protein of 116 kDa and approaches it at 4% gel density. Thus, at 9% SDS the apparent molecular mass is 180 kDa and extrapolates to 100 kDa at zero gel density. A similar slope difference is seen for the presumptive trimer, which crosses the line connecting the 200-kDa marker and approaches an extrapolated apparent molecular mass of 145 kDa. The same trend is apparent for the presumptive tetramer bands, but suitable markers were not available. It seems safe to conclude that the protein bands identified as presumptive dimers, timers, and tetramers are, in fact, likely to be just that. Similar size anomalies were noted in tubulin dimers cross-linked with hexanedione (Boekelheide, 1987) but not after cross-linking with a carbodiimide (Mejillano & Himes, 1991). It is, however, not clear at present whether or not the conformation of the dimers and trimers is the only factor contributing to the apparent high molecular masses. Analogues. A number of analogues of differing chemical properties bind at the colchicine binding site of tubulin. We have selected several of these absorbing beyond the protein spectrum to test the requirements for photosensitized dimerization (Figure 4). Thiocolchicine,which has a thiomethyl group in the 10-position, was less than one-tenth as potent in promoting covalent dimer formation as was colchicine (Figure 5). Although it binds well to tubulin (Chabin & Hastie, 1990), it fluoresces poorly, presumably because of rapid relaxation by nonradiative pathways (Figure 5). On the other hand, colchicide, lacking the 10-methoxy group (Figure 4), promoted photodimerization efficiently (Table I). Not surprisingly, colchicoside, which is sterically prevented from binding to the site, does not have significant antimitotic activity and fails to block colchicine binding (Zweig & Chignell, 1973), had virtually no ability ( colchicide > isocolchicine > 2-methoxy-5-nitrotropone 1 tropolone > thiocolchicine. The kinetics of colchicine binding to tubulin are complex and can be divided into two steps: a rapid, rather low-affinity binding reaction and a slow process believed to involve conformational changes in both colchicine and tubulin (Garland, 1978; Lambeir & Engelborghs, 12981; Baine et al., 1984; Engelborghs & Fitzgerald, 1987;Detrich et al., 1981;Brossi et al., 1988;Sackett et al., 1989). Because preincubation of colchicine with tubulin was not absolutely required to achieve photosensitized dimer fomration (Table I), the possibility existed that cross-linking could result from colchicine occupancy of the low-affinity component of the binding site represented by the rapid, first kinetic step. This hypothesis is strengthened by the finding that isocolchicine, which shows the low-affinity binding only (Hastie et al., 1989), also promoted tubulin dimerization when it was irradiated at the tropolone absorption maximum. That this was not simply nonspecific photosensitization was shown when the binding site was occupied by podophyllotoxin (Wolff et al., 1991; Cortese et al., 1977) which does not absorb at the tropolone wavelength but blocks colchicine binding. Under these conditions colchicine-induced photosensitization was virtually abolished, as was MTC-induced photosensitization. Moreover, since colchicoside did not photosensitize dimerization,it is clear that the mere presence of a colchicine analogue in solution (Le., not bound) is insufficient to promote the process. Together, these data provide evidence that occupancy of that part of the colchicine binding site involved in the rapid, low-affinity kinetic step for the binding of colchicine, MTC, or isocolchicine is sufficient to photosensitize tubulin dimerization. ACKNOWLEDGMENTS We thank Dr. Hans Cahnmann for the use of his irradiator and Dr. B. Bhattacharyya for numerous helpful suggestions. Registry No. MTC,60423-21-4; colchicine, 64-86-8; isocolchicine, 5 18-12-7; thiocolchicine, 2730-71-4;tropolone, 533-75-5; colchicoside, 477-29-2; colchicide, 518-15-0;2-methoxy-5-nitrotropone, 14628-90-1. REFERENCES Amrhein, N., & Filner, P. (1973)FEBS Lett. 33, 139-142. Andreu, J. M., & Timasheff, S.N. (1982)Biochemistry 21, 533-543. Andrews, A. T. (1986)Electrophoresis, 2nd ed., Clarendon Press, Oxford. Baine, S . , Puett, D., MacDonald, T. L., & Williams, R. C., Jr. (1984)J . Biol. Chem. 259, 7391-7398. Bhattacharyya, B., & Wolff, J. (1974)Proc. Natl. Acad. Sci. U.S.A. 71, 2627-263 1. Boekelheide, K. (1987) Toxicol. Appl. Pharmacol. 88, 383-3 96. Borisy, G. G., Olmsted, J. B., & Klugman, R. A. (1972)Proc. Natl. Acad. Sci. U.S.A. 69, 2890-2894. Brossi, A., Yeh, H. J. C., Chrzanowska, M., Wolff, J., Hamel, E., Lin, C. M., Quin, F., Suffness, M., & Silverton, J. (1988) Med. Res. Rev. 88, 77-94. Bryan, J. (1975) Ann. N.Y. Acad. Sci. 253, 247-259. Chabin, R. M.,& Hastie, S.B. (1990)Biochem. Biophys. Res. Commun. 161, 544-550. Cortese, F., Bhattacharyya, B., & Wolff, J. (1977) J . Biol. Chem. 252, 1134-1 140. Detrich, H. W., Williams, R. C., MacDonald, T. L., Wilson, L., & Puett, D. (1981)Biochemistry 20, 5999-6005.
3940
Biochemistry 1992, 31, 3940-3946
Engelborghs, Y . (1981) J . Biol. Chem. 256, 3276-3278. Engelborghs, Y . , & Fitzgerald, T. J. (1987) J . Biol. Chem. 262, 5204-5209. Gallela, G., & Smith, D. (1982) Can. J . Biochem. 60, 71-80. Garland, D. L. (1978) Biochemistry 17, 4266-4272. Hamel, E., & Lin, C. M. (1984) Biochemistry 28,2662-2667. Hastie, S.B., Williams, R. C., Jr., Puett, D., & MacDonald, T. L. (1989) J . Biol. Chem. 264, 6682-6688. Kasha, M. (1949) J . Opt. SOC.Am. 38, 929-934. Lambeir, A,, & Engelborghs, Y . (1 98 1) J . Biol. Chem. 256, 3279-3282. Luduena, R. F., Shooter, E. M., & Wilson, L. (1977) J . Biol. Chem. 252, 7006-7014. Maity, S.N., & Bhattacharyya, B. (1987) FEBS Lett. 218, 102-106.
Mejillano, M. R., & Himes, R. H. (1991) J . Biol. Chem. 266, 657-664. Sackett, D. L., Zimmerman, D. A., & Wolff, J. (1989) Biochemistry 28, 2662-2667. Sioussat, T. M., & Boekelheide, K. (1989) Biochemistry 28, 4435-4443 * Watanabe, T., & Flavin, M. (1973) Fed. Proc. 32, 642. Wilson, L., & Friedkin, M. (1966) Biochemistry 5, 2463-2468. Wolff, J., Knipling, L., Cahnmann, H. J. & Palumbo, G. (1991) Proc. Natl. Acad. Sci. U.S.A. 88, 2820-2824. Zaremba, T. G., LeBon, T. R., Millar, D. B., Funejkal, R. M., & Hawley, R. J. (1984) Biochemistry 23, 1073-1080. Zweig, M. H., & Chignell, C. F. (1973) Biochem. Pharmacol. 23, 2142-T973.
Errors in RNA NOESY Distance Measurements in Chimeric and Hybrid Duplexes: Differences in RNA and DNA Proton Relaxation? Andy C. Wang,t Seong G. Kim,gJ Peter F. Flynn,gJ Shan-Ho Chou,# John Orban,l!v and Brian R. Reid*J,s Biochemistry Department, Chemistry Department, and Howard Hughes Medical Institute, University of Washington, Seattle, Washington 981 95 Received November 15, 1991; Revised Manuscript Received February 10, 1992
ABSTRACT: Nuclear magnetic resonance experiments reveal that the base H8/H6 protons of oligoribonucleotides (RNA) have TI relaxation times that are distinctly longer than those of oligodeoxyribonucleotides (DNA). Similarly, the TIvalues for the R N A H1’protons are approximately twice those of the corresponding DNA H1’protons. These relaxation differences persist in single duplexes containing covalently linked RNA and DNA segments and cause serious overestimation of distances involving RNA protons in typical NOESY spectra collected with a duty cycle of 2-3 s. N M R and circular dichroism experiments indicate that the segments of R N A maintain their A-form geometry even in the interior of DNA-RNA-DNA chimeric duplexes, suggesting that the relaxation times are correlated with the type of helix topology. The difference in local proton density is the major cause of the longer nonselective T p of R N A compared to DNA, although small differences in internal motion cannot be completely ruled out. Fortunately, any internal motion differences that might exist are shown to be too small to affect cross-relaxation rates, and therefore reliable distance data can be obtained from time-dependent NOESY data sets provided an adequately long relaxation delay is used. In hybrid or chimeric RNA-DNA duplexes, if the longer R N A relaxation times are not taken into account in the recycle delay of NOESY pulse sequences, serious errors in measuring R N A proton distances are introduced.
M o d e r n molecular biology research has greatly expanded our knowledge of RNA function in recent years (Hoffman, 1991). Like proteins, RNA performs a wide variety of structural and functional roles, ranging from transfer RNA ‘This investigation was supported by Grants GM32681 and GM42896, awarded by the National Institute of Health, U S . Public Health Service. * Address correspondence to this author. *Biochemistry Department, University of Washington. Chemistry Department, University of Washington. Present address: Center for Magnetic Resonance, University of Minnesota, Minneapolis, MN 55455. Present address: Hare Research Incorporated, 18943 120th Ave. N.E., Suite 104, Bothell, WA 98011. # Howard Hughes Medical Institute, University of Washington. Present address: Center for Advanced Research in Biotechnology, Rockville, MD 20850.
*
molecules, with their distinctive cloverleaf structures, to the replicating “killer toxin”-encodingdoublestranded yeast RNA (Wickner et al., 1986). In addition, RNA carries the genetic transcript as messenger RNA, is found in association with proteins [e.g., ribosomes, spliceosomes, and RNaseP (Guerrier-Takada et al., 1983)], and has been found to self-splice and act as a catalyst (Cech & Bass, 1986). It also plays a role in gene regulation as antisense RNA (Green et al., 1986) and is the genetic material of many families of viruses (e.g., retroviruses). Recent advances in solid-state phosphoramidite chemistry have made it possible to routinely synthesize very pure RNA oligomers (Ogilvie et al., 1988) in NMR quantities (Chou et al., 1989). Thus defined sequence DNA-RNA hybrids are now amenable to structural investigation using NMR techniques. Indeed, because the strategies for DNA and RNA synthesis are compatible, one can study not only
0006-2960/92/043 1-3940$03.00/0 0 1992 American Chemical Society
